DuoNeural Cosmos3-Nano abliterated: 2-pass PRISM, layers 15-32, 201 weights (24.7%)
Browse files- .gitattributes +1 -0
- README.md +95 -0
- model_index.json +29 -0
- scheduler/scheduler_config.json +33 -0
- sound_tokenizer/config.json +67 -0
- sound_tokenizer/diffusion_pytorch_model.safetensors +3 -0
- text_tokenizer/chat_template.jinja +120 -0
- text_tokenizer/tokenizer.json +3 -0
- text_tokenizer/tokenizer_config.json +15 -0
- transformer/config.json +60 -0
- transformer/diffusion_pytorch_model-00001-of-00004.safetensors +3 -0
- transformer/diffusion_pytorch_model-00002-of-00004.safetensors +3 -0
- transformer/diffusion_pytorch_model-00003-of-00004.safetensors +3 -0
- transformer/diffusion_pytorch_model-00004-of-00004.safetensors +3 -0
- transformer/diffusion_pytorch_model.safetensors.index.json +816 -0
- vae/config.json +129 -0
- vae/diffusion_pytorch_model.safetensors +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
text_tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,95 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Cosmos3-Nano-Abliterated
|
| 2 |
+
|
| 3 |
+
**DuoNeural Research Lab | 2026-06-02**
|
| 4 |
+
|
| 5 |
+
## Model Description
|
| 6 |
+
|
| 7 |
+
`Cosmos3-Nano-Abliterated` is a refusal-projection-abliterated version of [NVIDIA/Cosmos3-Nano](https://huggingface.co/nvidia/Cosmos3-Nano), produced using DuoNeural's 2-pass PRISM abliteration methodology targeting the autoregressive understanding (`und_seq`) pathway of the Mixture-of-Transformers architecture.
|
| 8 |
+
|
| 9 |
+
**Base model:** `nvidia/Cosmos3-Nano` (15.16B parameters, MoT architecture)
|
| 10 |
+
**Method:** 2-pass weight-space abliteration (4-bit GPU for residuals, BF16 CPU for weight modification)
|
| 11 |
+
**Target:** Late-layer AR pathway weights — layers 15–32 of the 36-layer transformer
|
| 12 |
+
**Intended use:** Unconstrained video generation research, safety mechanism study
|
| 13 |
+
|
| 14 |
+
## Architecture Notes
|
| 15 |
+
|
| 16 |
+
Cosmos3-Nano uses a **Mixture-of-Transformers (MoT)** architecture with two parallel processing streams:
|
| 17 |
+
- `und_seq`: text/understanding AR pathway (conditions on prompt)
|
| 18 |
+
- `gen_seq`: video DiT pathway (denoises latent frames)
|
| 19 |
+
|
| 20 |
+
Both streams cross-attend in each of 36 transformer layers via joint attention (`add_q_proj`, `add_k_proj`, `add_v_proj`, `to_add_out`). The abliteration targets the AR pathway's late-layer projection weights where safety-relevant activations are concentrated.
|
| 21 |
+
|
| 22 |
+
**Note:** The action prediction head (`action_modality_embed`, `action_proj_in/out`) is absent from the Nano variant. Cosmos3-Nano supports video generation only; robot action output requires the full Cosmos3 model.
|
| 23 |
+
|
| 24 |
+
## Abliteration Details
|
| 25 |
+
|
| 26 |
+
| Parameter | Value |
|
| 27 |
+
|-----------|-------|
|
| 28 |
+
| Method | 2-pass PRISM abliteration |
|
| 29 |
+
| Total weights modified | 201 / 814 (24.7%) |
|
| 30 |
+
| Primary target region | Layers 15–32 (AR pathway late layers) |
|
| 31 |
+
| Primary region tensors modified | 137 tensors |
|
| 32 |
+
| Primary region mean relative change | 6.54% |
|
| 33 |
+
| Control region (layers 0–14) bleedthrough | 63 tensors, mean 1.66% |
|
| 34 |
+
| Max relative change (layer 32 joint attn norms) | 100% (replaced) |
|
| 35 |
+
| Pass 1 | 4-bit GPU: residual direction projection |
|
| 36 |
+
| Pass 2 | BF16 CPU: weight modification via refusal direction subtraction |
|
| 37 |
+
|
| 38 |
+
### Layer-Level Summary
|
| 39 |
+
|
| 40 |
+
Layer 32 sustained the most aggressive modifications — the joint attention normalization weights (`norm_added_k`, `norm_added_q`, `norm_k`, `norm_q`) and `add_k_proj`/`add_v_proj` projections were completely replaced (rel=1.0). This corresponds to a late cross-modal routing point where safety-conditioned refusal signals concentrate in the MoT AR pathway.
|
| 41 |
+
|
| 42 |
+
| Region | Layers affected | Mean Δ | Max Δ |
|
| 43 |
+
|--------|----------------|--------|-------|
|
| 44 |
+
| Abliterated (15–35) | 15–28, 32 | 6.54% | 100% |
|
| 45 |
+
| Control (0–14) bleedthrough | 8–14 | 1.66% | 3.62% |
|
| 46 |
+
|
| 47 |
+
### Quality Evaluation Note
|
| 48 |
+
|
| 49 |
+
Standard KL divergence methodology (Heretic v2: full vocab, first-token logits) is **not applicable** to Cosmos3-Nano's diffusion architecture — the model requires joint `und_seq` + `gen_seq` forward passes and does not expose standalone text token logits. Weight-level metrics are provided above as a proxy. Full quality evaluation requires comparing generated video distributions via the complete Cosmos3OmniPipeline.
|
| 50 |
+
|
| 51 |
+
## Usage
|
| 52 |
+
|
| 53 |
+
```python
|
| 54 |
+
from diffusers import Cosmos3OmniPipeline
|
| 55 |
+
import torch
|
| 56 |
+
|
| 57 |
+
pipe = Cosmos3OmniPipeline.from_pretrained(
|
| 58 |
+
"DuoNeural/Cosmos3-Nano-Abliterated",
|
| 59 |
+
torch_dtype=torch.bfloat16,
|
| 60 |
+
)
|
| 61 |
+
pipe = pipe.to("cuda")
|
| 62 |
+
|
| 63 |
+
output = pipe(
|
| 64 |
+
prompt="Your prompt here",
|
| 65 |
+
num_frames=16,
|
| 66 |
+
height=256,
|
| 67 |
+
width=256,
|
| 68 |
+
guidance_scale=7.0,
|
| 69 |
+
)
|
| 70 |
+
```
|
| 71 |
+
|
| 72 |
+
**Requirements:** ~48GB VRAM (A100 80GB recommended). BF16. CUDA only.
|
| 73 |
+
|
| 74 |
+
## Ethical Statement
|
| 75 |
+
|
| 76 |
+
Released for research purposes: studying safety mechanisms in omni-modal AI, abliteration methodology development, and unconstrained video generation research. DuoNeural publishes both abliterated models and methodology openly to advance scientific understanding of post-training safety interventions.
|
| 77 |
+
|
| 78 |
+
---
|
| 79 |
+
|
| 80 |
+
## About DuoNeural
|
| 81 |
+
|
| 82 |
+
**DuoNeural** is an open AI research lab at the intersection of human and artificial intelligence. 30+ open-access papers, 69+ HuggingFace models, experiments on consumer GPUs and real QPUs.
|
| 83 |
+
|
| 84 |
+
### Selected Papers
|
| 85 |
+
- **P4** — The Dynamical Horizon Principle. [doi.org/10.5281/zenodo.19952612](https://doi.org/10.5281/zenodo.19952612)
|
| 86 |
+
- **P28** — DHP is a Recurrence Constraint (includes Cosmos3 physical DHP probe). [doi.org/10.5281/zenodo.20476068](https://doi.org/10.5281/zenodo.20476068)
|
| 87 |
+
|
| 88 |
+
### Team
|
| 89 |
+
| Member | Role |
|
| 90 |
+
|--------|------|
|
| 91 |
+
| Jesse Caldwell | Founder |
|
| 92 |
+
| Archon | Lab Director — post-training, abliteration, quantum |
|
| 93 |
+
| Aura | Research AI — synthesis, red-teaming |
|
| 94 |
+
|
| 95 |
+
🤗 [DuoNeural](https://huggingface.co/DuoNeural) | 🌐 [duoneural.com](https://duoneural.com) | 📚 [zenodo.org/communities/duoneural](https://zenodo.org/communities/duoneural) | 📧 duoneural@proton.me
|
model_index.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "Cosmos3OmniPipeline",
|
| 3 |
+
"_diffusers_version": "0.39.0.dev0",
|
| 4 |
+
"_name_or_path": "/workspace/cosmos3-nano",
|
| 5 |
+
"safety_checker": [
|
| 6 |
+
null,
|
| 7 |
+
null
|
| 8 |
+
],
|
| 9 |
+
"scheduler": [
|
| 10 |
+
"diffusers",
|
| 11 |
+
"UniPCMultistepScheduler"
|
| 12 |
+
],
|
| 13 |
+
"sound_tokenizer": [
|
| 14 |
+
"diffusers",
|
| 15 |
+
"Cosmos3AVAEAudioTokenizer"
|
| 16 |
+
],
|
| 17 |
+
"text_tokenizer": [
|
| 18 |
+
"transformers",
|
| 19 |
+
"Qwen2Tokenizer"
|
| 20 |
+
],
|
| 21 |
+
"transformer": [
|
| 22 |
+
"diffusers",
|
| 23 |
+
"Cosmos3OmniTransformer"
|
| 24 |
+
],
|
| 25 |
+
"vae": [
|
| 26 |
+
"diffusers",
|
| 27 |
+
"AutoencoderKLWan"
|
| 28 |
+
]
|
| 29 |
+
}
|
scheduler/scheduler_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "UniPCMultistepScheduler",
|
| 3 |
+
"_diffusers_version": "0.39.0.dev0",
|
| 4 |
+
"beta_end": 0.02,
|
| 5 |
+
"beta_schedule": "linear",
|
| 6 |
+
"beta_start": 0.0001,
|
| 7 |
+
"disable_corrector": [],
|
| 8 |
+
"dynamic_thresholding_ratio": 0.995,
|
| 9 |
+
"final_sigmas_type": "zero",
|
| 10 |
+
"flow_shift": 1.0,
|
| 11 |
+
"lower_order_final": true,
|
| 12 |
+
"num_train_timesteps": 1000,
|
| 13 |
+
"predict_x0": true,
|
| 14 |
+
"prediction_type": "flow_prediction",
|
| 15 |
+
"rescale_betas_zero_snr": false,
|
| 16 |
+
"sample_max_value": 1.0,
|
| 17 |
+
"shift_terminal": null,
|
| 18 |
+
"sigma_max": 200.0,
|
| 19 |
+
"sigma_min": 0.147,
|
| 20 |
+
"solver_order": 2,
|
| 21 |
+
"solver_p": null,
|
| 22 |
+
"solver_type": "bh2",
|
| 23 |
+
"steps_offset": 0,
|
| 24 |
+
"thresholding": false,
|
| 25 |
+
"time_shift_type": "exponential",
|
| 26 |
+
"timestep_spacing": "linspace",
|
| 27 |
+
"trained_betas": null,
|
| 28 |
+
"use_beta_sigmas": false,
|
| 29 |
+
"use_dynamic_shifting": false,
|
| 30 |
+
"use_exponential_sigmas": false,
|
| 31 |
+
"use_flow_sigmas": true,
|
| 32 |
+
"use_karras_sigmas": true
|
| 33 |
+
}
|
sound_tokenizer/config.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "Cosmos3AVAEAudioTokenizer",
|
| 3 |
+
"_diffusers_version": "0.39.0.dev0",
|
| 4 |
+
"_name_or_path": "/workspace/cosmos3-nano/sound_tokenizer",
|
| 5 |
+
"activation": "snakebeta",
|
| 6 |
+
"anti_aliasing": false,
|
| 7 |
+
"bottleneck": {
|
| 8 |
+
"type": "vae"
|
| 9 |
+
},
|
| 10 |
+
"bottleneck_type": "vae",
|
| 11 |
+
"causal": false,
|
| 12 |
+
"dec_anti_aliasing": false,
|
| 13 |
+
"dec_c_mults": [
|
| 14 |
+
1,
|
| 15 |
+
2,
|
| 16 |
+
4,
|
| 17 |
+
8,
|
| 18 |
+
16
|
| 19 |
+
],
|
| 20 |
+
"dec_dim": 320,
|
| 21 |
+
"dec_final_tanh": false,
|
| 22 |
+
"dec_out_channels": 2,
|
| 23 |
+
"dec_strides": [
|
| 24 |
+
2,
|
| 25 |
+
4,
|
| 26 |
+
5,
|
| 27 |
+
6,
|
| 28 |
+
8
|
| 29 |
+
],
|
| 30 |
+
"dec_type": "oobleck",
|
| 31 |
+
"dec_use_nearest_upsample": false,
|
| 32 |
+
"dec_use_snake": true,
|
| 33 |
+
"dec_use_tanh_at_final": false,
|
| 34 |
+
"enc_c_mults": [
|
| 35 |
+
1,
|
| 36 |
+
2,
|
| 37 |
+
4
|
| 38 |
+
],
|
| 39 |
+
"enc_dim": 192,
|
| 40 |
+
"enc_hop_length": 16,
|
| 41 |
+
"enc_identity_init": false,
|
| 42 |
+
"enc_intermediate_dim": 768,
|
| 43 |
+
"enc_latent_dim": 128,
|
| 44 |
+
"enc_n_fft": 64,
|
| 45 |
+
"enc_num_blocks": 2,
|
| 46 |
+
"enc_num_layers": 12,
|
| 47 |
+
"enc_strides": [
|
| 48 |
+
4,
|
| 49 |
+
5,
|
| 50 |
+
6
|
| 51 |
+
],
|
| 52 |
+
"enc_type": "spec_convnext",
|
| 53 |
+
"enc_use_snake": true,
|
| 54 |
+
"hop_size": 1920,
|
| 55 |
+
"input_channels": 1,
|
| 56 |
+
"latent_mean": null,
|
| 57 |
+
"latent_std": null,
|
| 58 |
+
"model_type": "autoencoder_v2",
|
| 59 |
+
"normalize_volume": true,
|
| 60 |
+
"padding_mode": "zeros",
|
| 61 |
+
"sampling_rate": 48000,
|
| 62 |
+
"snake_logscale": true,
|
| 63 |
+
"stereo": true,
|
| 64 |
+
"use_cuda_kernel": false,
|
| 65 |
+
"use_wav_as_input": true,
|
| 66 |
+
"vocoder_input_dim": 64
|
| 67 |
+
}
|
sound_tokenizer/diffusion_pytorch_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:595a7ccb1b0de3c0f53c348a9667d5e7b0cd2089b4315c3544c4188ec2569895
|
| 3 |
+
size 946214860
|
text_tokenizer/chat_template.jinja
ADDED
|
@@ -0,0 +1,120 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{%- if messages[0].content is string %}
|
| 5 |
+
{{- messages[0].content }}
|
| 6 |
+
{%- else %}
|
| 7 |
+
{%- for content in messages[0].content %}
|
| 8 |
+
{%- if 'text' in content %}
|
| 9 |
+
{{- content.text }}
|
| 10 |
+
{%- endif %}
|
| 11 |
+
{%- endfor %}
|
| 12 |
+
{%- endif %}
|
| 13 |
+
{{- '\n\n' }}
|
| 14 |
+
{%- endif %}
|
| 15 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 16 |
+
{%- for tool in tools %}
|
| 17 |
+
{{- "\n" }}
|
| 18 |
+
{{- tool | tojson }}
|
| 19 |
+
{%- endfor %}
|
| 20 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 21 |
+
{%- else %}
|
| 22 |
+
{%- if messages[0].role == 'system' %}
|
| 23 |
+
{{- '<|im_start|>system\n' }}
|
| 24 |
+
{%- if messages[0].content is string %}
|
| 25 |
+
{{- messages[0].content }}
|
| 26 |
+
{%- else %}
|
| 27 |
+
{%- for content in messages[0].content %}
|
| 28 |
+
{%- if 'text' in content %}
|
| 29 |
+
{{- content.text }}
|
| 30 |
+
{%- endif %}
|
| 31 |
+
{%- endfor %}
|
| 32 |
+
{%- endif %}
|
| 33 |
+
{{- '<|im_end|>\n' }}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{%- endif %}
|
| 36 |
+
{%- set image_count = namespace(value=0) %}
|
| 37 |
+
{%- set video_count = namespace(value=0) %}
|
| 38 |
+
{%- for message in messages %}
|
| 39 |
+
{%- if message.role == "user" %}
|
| 40 |
+
{{- '<|im_start|>' + message.role + '\n' }}
|
| 41 |
+
{%- if message.content is string %}
|
| 42 |
+
{{- message.content }}
|
| 43 |
+
{%- else %}
|
| 44 |
+
{%- for content in message.content %}
|
| 45 |
+
{%- if content.type == 'image' or 'image' in content or 'image_url' in content %}
|
| 46 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 47 |
+
{%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}
|
| 48 |
+
<|vision_start|><|image_pad|><|vision_end|>
|
| 49 |
+
{%- elif content.type == 'video' or 'video' in content %}
|
| 50 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 51 |
+
{%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}
|
| 52 |
+
<|vision_start|><|video_pad|><|vision_end|>
|
| 53 |
+
{%- elif 'text' in content %}
|
| 54 |
+
{{- content.text }}
|
| 55 |
+
{%- endif %}
|
| 56 |
+
{%- endfor %}
|
| 57 |
+
{%- endif %}
|
| 58 |
+
{{- '<|im_end|>\n' }}
|
| 59 |
+
{%- elif message.role == "assistant" %}
|
| 60 |
+
{{- '<|im_start|>' + message.role + '\n' }}
|
| 61 |
+
{%- if message.content is string %}
|
| 62 |
+
{{- message.content }}
|
| 63 |
+
{%- else %}
|
| 64 |
+
{%- for content_item in message.content %}
|
| 65 |
+
{%- if 'text' in content_item %}
|
| 66 |
+
{{- content_item.text }}
|
| 67 |
+
{%- endif %}
|
| 68 |
+
{%- endfor %}
|
| 69 |
+
{%- endif %}
|
| 70 |
+
{%- if message.tool_calls %}
|
| 71 |
+
{%- for tool_call in message.tool_calls %}
|
| 72 |
+
{%- if (loop.first and message.content) or (not loop.first) %}
|
| 73 |
+
{{- '\n' }}
|
| 74 |
+
{%- endif %}
|
| 75 |
+
{%- if tool_call.function %}
|
| 76 |
+
{%- set tool_call = tool_call.function %}
|
| 77 |
+
{%- endif %}
|
| 78 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 79 |
+
{{- tool_call.name }}
|
| 80 |
+
{{- '", "arguments": ' }}
|
| 81 |
+
{%- if tool_call.arguments is string %}
|
| 82 |
+
{{- tool_call.arguments }}
|
| 83 |
+
{%- else %}
|
| 84 |
+
{{- tool_call.arguments | tojson }}
|
| 85 |
+
{%- endif %}
|
| 86 |
+
{{- '}\n</tool_call>' }}
|
| 87 |
+
{%- endfor %}
|
| 88 |
+
{%- endif %}
|
| 89 |
+
{{- '<|im_end|>\n' }}
|
| 90 |
+
{%- elif message.role == "tool" %}
|
| 91 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 92 |
+
{{- '<|im_start|>user' }}
|
| 93 |
+
{%- endif %}
|
| 94 |
+
{{- '\n<tool_response>\n' }}
|
| 95 |
+
{%- if message.content is string %}
|
| 96 |
+
{{- message.content }}
|
| 97 |
+
{%- else %}
|
| 98 |
+
{%- for content in message.content %}
|
| 99 |
+
{%- if content.type == 'image' or 'image' in content or 'image_url' in content %}
|
| 100 |
+
{%- set image_count.value = image_count.value + 1 %}
|
| 101 |
+
{%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}
|
| 102 |
+
<|vision_start|><|image_pad|><|vision_end|>
|
| 103 |
+
{%- elif content.type == 'video' or 'video' in content %}
|
| 104 |
+
{%- set video_count.value = video_count.value + 1 %}
|
| 105 |
+
{%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}
|
| 106 |
+
<|vision_start|><|video_pad|><|vision_end|>
|
| 107 |
+
{%- elif 'text' in content %}
|
| 108 |
+
{{- content.text }}
|
| 109 |
+
{%- endif %}
|
| 110 |
+
{%- endfor %}
|
| 111 |
+
{%- endif %}
|
| 112 |
+
{{- '\n</tool_response>' }}
|
| 113 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 114 |
+
{{- '<|im_end|>\n' }}
|
| 115 |
+
{%- endif %}
|
| 116 |
+
{%- endif %}
|
| 117 |
+
{%- endfor %}
|
| 118 |
+
{%- if add_generation_prompt %}
|
| 119 |
+
{{- '<|im_start|>assistant\n' }}
|
| 120 |
+
{%- endif %}
|
text_tokenizer/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
| 3 |
+
size 11422650
|
text_tokenizer/tokenizer_config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"is_local": true,
|
| 9 |
+
"local_files_only": false,
|
| 10 |
+
"model_max_length": 262144,
|
| 11 |
+
"pad_token": "<|endoftext|>",
|
| 12 |
+
"split_special_tokens": false,
|
| 13 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 14 |
+
"unk_token": null
|
| 15 |
+
}
|
transformer/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "Cosmos3OmniTransformer",
|
| 3 |
+
"_diffusers_version": "0.39.0.dev0",
|
| 4 |
+
"_name_or_path": "/workspace/cosmos3-nano/transformer",
|
| 5 |
+
"action_dim": 64,
|
| 6 |
+
"action_gen": true,
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"base_fps": 24,
|
| 10 |
+
"dtype": "bfloat16",
|
| 11 |
+
"enable_fps_modulation": true,
|
| 12 |
+
"freeze_und": false,
|
| 13 |
+
"head_dim": 128,
|
| 14 |
+
"hidden_act": "silu",
|
| 15 |
+
"hidden_size": 4096,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 12288,
|
| 18 |
+
"joint_attn_implementation": "two_way",
|
| 19 |
+
"latent_channel": 48,
|
| 20 |
+
"latent_patch_size": 2,
|
| 21 |
+
"max_action_dim": 64,
|
| 22 |
+
"max_position_embeddings": 262144,
|
| 23 |
+
"model_type": "qwen3_vl_text",
|
| 24 |
+
"num_attention_heads": 32,
|
| 25 |
+
"num_embodiment_domains": 32,
|
| 26 |
+
"num_hidden_layers": 36,
|
| 27 |
+
"num_key_value_heads": 8,
|
| 28 |
+
"patch_latent_dim": 192,
|
| 29 |
+
"position_embedding_type": "unified_3d_mrope",
|
| 30 |
+
"qk_norm": false,
|
| 31 |
+
"qk_norm_for_diffusion": true,
|
| 32 |
+
"qk_norm_for_text": true,
|
| 33 |
+
"rms_norm_eps": 1e-06,
|
| 34 |
+
"rope_axes_dim": [
|
| 35 |
+
24,
|
| 36 |
+
20,
|
| 37 |
+
20
|
| 38 |
+
],
|
| 39 |
+
"rope_scaling": {
|
| 40 |
+
"mrope_interleaved": true,
|
| 41 |
+
"mrope_section": [
|
| 42 |
+
24,
|
| 43 |
+
20,
|
| 44 |
+
20
|
| 45 |
+
],
|
| 46 |
+
"rope_type": "default"
|
| 47 |
+
},
|
| 48 |
+
"rope_theta": 5000000,
|
| 49 |
+
"sound_dim": 64,
|
| 50 |
+
"sound_gen": true,
|
| 51 |
+
"sound_latent_fps": 25,
|
| 52 |
+
"temporal_compression_factor_sound": 1,
|
| 53 |
+
"timestep_scale": 0.001,
|
| 54 |
+
"unified_3d_mrope_reset_spatial_ids": true,
|
| 55 |
+
"unified_3d_mrope_temporal_modality_margin": 15000,
|
| 56 |
+
"use_cache": true,
|
| 57 |
+
"use_moe": true,
|
| 58 |
+
"video_temporal_causal": false,
|
| 59 |
+
"vocab_size": 151936
|
| 60 |
+
}
|
transformer/diffusion_pytorch_model-00001-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa8c3ccb582d6a8f4bd4d1e7962a05300d942939ed2a43f4144306f5859f3aab
|
| 3 |
+
size 9902112712
|
transformer/diffusion_pytorch_model-00002-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:32bb1fc2f3fe500cfb69b2bb2dc5a1f81117247b971b7036524dbaefc54ba9a5
|
| 3 |
+
size 9999691408
|
transformer/diffusion_pytorch_model-00003-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5dd10e0db7a413d736d3084caf82c0f1f351093c30168912b9d02a8a605dd91
|
| 3 |
+
size 9127255800
|
transformer/diffusion_pytorch_model-00004-of-00004.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7ae0e0ba709dad404dcc4fefd76306c61099f5d7eea6d5647025c56ac9c1a64
|
| 3 |
+
size 1320208080
|
transformer/diffusion_pytorch_model.safetensors.index.json
ADDED
|
@@ -0,0 +1,816 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 30349177344
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"audio_modality_embed": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 7 |
+
"audio_proj_in.bias": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 8 |
+
"audio_proj_in.weight": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 9 |
+
"audio_proj_out.bias": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 10 |
+
"audio_proj_out.weight": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 11 |
+
"embed_tokens.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 12 |
+
"layers.0.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 13 |
+
"layers.0.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 14 |
+
"layers.0.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 15 |
+
"layers.0.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 16 |
+
"layers.0.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 17 |
+
"layers.0.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 18 |
+
"layers.0.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 19 |
+
"layers.0.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 20 |
+
"layers.0.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 21 |
+
"layers.0.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 22 |
+
"layers.0.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 23 |
+
"layers.0.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 24 |
+
"layers.0.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 25 |
+
"layers.0.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 26 |
+
"layers.0.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 27 |
+
"layers.0.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 28 |
+
"layers.0.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 29 |
+
"layers.0.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 30 |
+
"layers.0.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 31 |
+
"layers.0.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 32 |
+
"layers.0.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 33 |
+
"layers.0.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 34 |
+
"layers.1.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 35 |
+
"layers.1.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 36 |
+
"layers.1.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 37 |
+
"layers.1.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 38 |
+
"layers.1.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 39 |
+
"layers.1.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 40 |
+
"layers.1.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 41 |
+
"layers.1.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 42 |
+
"layers.1.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 43 |
+
"layers.1.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 44 |
+
"layers.1.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 45 |
+
"layers.1.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 46 |
+
"layers.1.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 47 |
+
"layers.1.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 48 |
+
"layers.1.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 49 |
+
"layers.1.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 50 |
+
"layers.1.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 51 |
+
"layers.1.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 52 |
+
"layers.1.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 53 |
+
"layers.1.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 54 |
+
"layers.1.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 55 |
+
"layers.1.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 56 |
+
"layers.10.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 57 |
+
"layers.10.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 58 |
+
"layers.10.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 59 |
+
"layers.10.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 60 |
+
"layers.10.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 61 |
+
"layers.10.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 62 |
+
"layers.10.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 63 |
+
"layers.10.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 64 |
+
"layers.10.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 65 |
+
"layers.10.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 66 |
+
"layers.10.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 67 |
+
"layers.10.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 68 |
+
"layers.10.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 69 |
+
"layers.10.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 70 |
+
"layers.10.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 71 |
+
"layers.10.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 72 |
+
"layers.10.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 73 |
+
"layers.10.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 74 |
+
"layers.10.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 75 |
+
"layers.10.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 76 |
+
"layers.10.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 77 |
+
"layers.10.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 78 |
+
"layers.11.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 79 |
+
"layers.11.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 80 |
+
"layers.11.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 81 |
+
"layers.11.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 82 |
+
"layers.11.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 83 |
+
"layers.11.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 84 |
+
"layers.11.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 85 |
+
"layers.11.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 86 |
+
"layers.11.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 87 |
+
"layers.11.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 88 |
+
"layers.11.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 89 |
+
"layers.11.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 90 |
+
"layers.11.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 91 |
+
"layers.11.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 92 |
+
"layers.11.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 93 |
+
"layers.11.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 94 |
+
"layers.11.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 95 |
+
"layers.11.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 96 |
+
"layers.11.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 97 |
+
"layers.11.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 98 |
+
"layers.11.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 99 |
+
"layers.11.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 100 |
+
"layers.12.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 101 |
+
"layers.12.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 102 |
+
"layers.12.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 103 |
+
"layers.12.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 104 |
+
"layers.12.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 105 |
+
"layers.12.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 106 |
+
"layers.12.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 107 |
+
"layers.12.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 108 |
+
"layers.12.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 109 |
+
"layers.12.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 110 |
+
"layers.12.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 111 |
+
"layers.12.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 112 |
+
"layers.12.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 113 |
+
"layers.12.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 114 |
+
"layers.12.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 115 |
+
"layers.12.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 116 |
+
"layers.12.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 117 |
+
"layers.12.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 118 |
+
"layers.12.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 119 |
+
"layers.12.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 120 |
+
"layers.12.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 121 |
+
"layers.12.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 122 |
+
"layers.13.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 123 |
+
"layers.13.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 124 |
+
"layers.13.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 125 |
+
"layers.13.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 126 |
+
"layers.13.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 127 |
+
"layers.13.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 128 |
+
"layers.13.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 129 |
+
"layers.13.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 130 |
+
"layers.13.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 131 |
+
"layers.13.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 132 |
+
"layers.13.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 133 |
+
"layers.13.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 134 |
+
"layers.13.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 135 |
+
"layers.13.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 136 |
+
"layers.13.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 137 |
+
"layers.13.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 138 |
+
"layers.13.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 139 |
+
"layers.13.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 140 |
+
"layers.13.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 141 |
+
"layers.13.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 142 |
+
"layers.13.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 143 |
+
"layers.13.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 144 |
+
"layers.14.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 145 |
+
"layers.14.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 146 |
+
"layers.14.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 147 |
+
"layers.14.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 148 |
+
"layers.14.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 149 |
+
"layers.14.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 150 |
+
"layers.14.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 151 |
+
"layers.14.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 152 |
+
"layers.14.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 153 |
+
"layers.14.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 154 |
+
"layers.14.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 155 |
+
"layers.14.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 156 |
+
"layers.14.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 157 |
+
"layers.14.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 158 |
+
"layers.14.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 159 |
+
"layers.14.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 160 |
+
"layers.14.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 161 |
+
"layers.14.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 162 |
+
"layers.14.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 163 |
+
"layers.14.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 164 |
+
"layers.14.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 165 |
+
"layers.14.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 166 |
+
"layers.15.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 167 |
+
"layers.15.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 168 |
+
"layers.15.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 169 |
+
"layers.15.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 170 |
+
"layers.15.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 171 |
+
"layers.15.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 172 |
+
"layers.15.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 173 |
+
"layers.15.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 174 |
+
"layers.15.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 175 |
+
"layers.15.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 176 |
+
"layers.15.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 177 |
+
"layers.15.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 178 |
+
"layers.15.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 179 |
+
"layers.15.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 180 |
+
"layers.15.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 181 |
+
"layers.15.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 182 |
+
"layers.15.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 183 |
+
"layers.15.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 184 |
+
"layers.15.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 185 |
+
"layers.15.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 186 |
+
"layers.15.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 187 |
+
"layers.15.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 188 |
+
"layers.16.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 189 |
+
"layers.16.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 190 |
+
"layers.16.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 191 |
+
"layers.16.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 192 |
+
"layers.16.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 193 |
+
"layers.16.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 194 |
+
"layers.16.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 195 |
+
"layers.16.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 196 |
+
"layers.16.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 197 |
+
"layers.16.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 198 |
+
"layers.16.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 199 |
+
"layers.16.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 200 |
+
"layers.16.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 201 |
+
"layers.16.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 202 |
+
"layers.16.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 203 |
+
"layers.16.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 204 |
+
"layers.16.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 205 |
+
"layers.16.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 206 |
+
"layers.16.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 207 |
+
"layers.16.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 208 |
+
"layers.16.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 209 |
+
"layers.16.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 210 |
+
"layers.17.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 211 |
+
"layers.17.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 212 |
+
"layers.17.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 213 |
+
"layers.17.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 214 |
+
"layers.17.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 215 |
+
"layers.17.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 216 |
+
"layers.17.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 217 |
+
"layers.17.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 218 |
+
"layers.17.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 219 |
+
"layers.17.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 220 |
+
"layers.17.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 221 |
+
"layers.17.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 222 |
+
"layers.17.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 223 |
+
"layers.17.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 224 |
+
"layers.17.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 225 |
+
"layers.17.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 226 |
+
"layers.17.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 227 |
+
"layers.17.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 228 |
+
"layers.17.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 229 |
+
"layers.17.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 230 |
+
"layers.17.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 231 |
+
"layers.17.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 232 |
+
"layers.18.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 233 |
+
"layers.18.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 234 |
+
"layers.18.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 235 |
+
"layers.18.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 236 |
+
"layers.18.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 237 |
+
"layers.18.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 238 |
+
"layers.18.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 239 |
+
"layers.18.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 240 |
+
"layers.18.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 241 |
+
"layers.18.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 242 |
+
"layers.18.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 243 |
+
"layers.18.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 244 |
+
"layers.18.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 245 |
+
"layers.18.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 246 |
+
"layers.18.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 247 |
+
"layers.18.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 248 |
+
"layers.18.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 249 |
+
"layers.18.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 250 |
+
"layers.18.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 251 |
+
"layers.18.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 252 |
+
"layers.18.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 253 |
+
"layers.18.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 254 |
+
"layers.19.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 255 |
+
"layers.19.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 256 |
+
"layers.19.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 257 |
+
"layers.19.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 258 |
+
"layers.19.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 259 |
+
"layers.19.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 260 |
+
"layers.19.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 261 |
+
"layers.19.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 262 |
+
"layers.19.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 263 |
+
"layers.19.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 264 |
+
"layers.19.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 265 |
+
"layers.19.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 266 |
+
"layers.19.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 267 |
+
"layers.19.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 268 |
+
"layers.19.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 269 |
+
"layers.19.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 270 |
+
"layers.19.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 271 |
+
"layers.19.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 272 |
+
"layers.19.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 273 |
+
"layers.19.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 274 |
+
"layers.19.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 275 |
+
"layers.19.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 276 |
+
"layers.2.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 277 |
+
"layers.2.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 278 |
+
"layers.2.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 279 |
+
"layers.2.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 280 |
+
"layers.2.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 281 |
+
"layers.2.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 282 |
+
"layers.2.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 283 |
+
"layers.2.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 284 |
+
"layers.2.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 285 |
+
"layers.2.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 286 |
+
"layers.2.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 287 |
+
"layers.2.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 288 |
+
"layers.2.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 289 |
+
"layers.2.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 290 |
+
"layers.2.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 291 |
+
"layers.2.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 292 |
+
"layers.2.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 293 |
+
"layers.2.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 294 |
+
"layers.2.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 295 |
+
"layers.2.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 296 |
+
"layers.2.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 297 |
+
"layers.2.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 298 |
+
"layers.20.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 299 |
+
"layers.20.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 300 |
+
"layers.20.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 301 |
+
"layers.20.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 302 |
+
"layers.20.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 303 |
+
"layers.20.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 304 |
+
"layers.20.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 305 |
+
"layers.20.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 306 |
+
"layers.20.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 307 |
+
"layers.20.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 308 |
+
"layers.20.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 309 |
+
"layers.20.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 310 |
+
"layers.20.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 311 |
+
"layers.20.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 312 |
+
"layers.20.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 313 |
+
"layers.20.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 314 |
+
"layers.20.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 315 |
+
"layers.20.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 316 |
+
"layers.20.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 317 |
+
"layers.20.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 318 |
+
"layers.20.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 319 |
+
"layers.20.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 320 |
+
"layers.21.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 321 |
+
"layers.21.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 322 |
+
"layers.21.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 323 |
+
"layers.21.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 324 |
+
"layers.21.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 325 |
+
"layers.21.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 326 |
+
"layers.21.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 327 |
+
"layers.21.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 328 |
+
"layers.21.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 329 |
+
"layers.21.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 330 |
+
"layers.21.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 331 |
+
"layers.21.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 332 |
+
"layers.21.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 333 |
+
"layers.21.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 334 |
+
"layers.21.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 335 |
+
"layers.21.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 336 |
+
"layers.21.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 337 |
+
"layers.21.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 338 |
+
"layers.21.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 339 |
+
"layers.21.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 340 |
+
"layers.21.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 341 |
+
"layers.21.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 342 |
+
"layers.22.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 343 |
+
"layers.22.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 344 |
+
"layers.22.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 345 |
+
"layers.22.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 346 |
+
"layers.22.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 347 |
+
"layers.22.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 348 |
+
"layers.22.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 349 |
+
"layers.22.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 350 |
+
"layers.22.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 351 |
+
"layers.22.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 352 |
+
"layers.22.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 353 |
+
"layers.22.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 354 |
+
"layers.22.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 355 |
+
"layers.22.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 356 |
+
"layers.22.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 357 |
+
"layers.22.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 358 |
+
"layers.22.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 359 |
+
"layers.22.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 360 |
+
"layers.22.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 361 |
+
"layers.22.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 362 |
+
"layers.22.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 363 |
+
"layers.22.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 364 |
+
"layers.23.input_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 365 |
+
"layers.23.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 366 |
+
"layers.23.mlp.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 367 |
+
"layers.23.mlp.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 368 |
+
"layers.23.mlp.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 369 |
+
"layers.23.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 370 |
+
"layers.23.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 371 |
+
"layers.23.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 372 |
+
"layers.23.post_attention_layernorm.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 373 |
+
"layers.23.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 374 |
+
"layers.23.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 375 |
+
"layers.23.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 376 |
+
"layers.23.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 377 |
+
"layers.23.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 378 |
+
"layers.23.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 379 |
+
"layers.23.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 380 |
+
"layers.23.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 381 |
+
"layers.23.self_attn.to_add_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 382 |
+
"layers.23.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 383 |
+
"layers.23.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 384 |
+
"layers.23.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 385 |
+
"layers.23.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 386 |
+
"layers.24.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 387 |
+
"layers.24.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 388 |
+
"layers.24.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 389 |
+
"layers.24.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 390 |
+
"layers.24.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 391 |
+
"layers.24.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 392 |
+
"layers.24.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 393 |
+
"layers.24.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 394 |
+
"layers.24.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 395 |
+
"layers.24.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 396 |
+
"layers.24.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 397 |
+
"layers.24.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 398 |
+
"layers.24.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 399 |
+
"layers.24.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 400 |
+
"layers.24.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 401 |
+
"layers.24.self_attn.norm_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 402 |
+
"layers.24.self_attn.norm_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 403 |
+
"layers.24.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 404 |
+
"layers.24.self_attn.to_k.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 405 |
+
"layers.24.self_attn.to_out.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 406 |
+
"layers.24.self_attn.to_q.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 407 |
+
"layers.24.self_attn.to_v.weight": "diffusion_pytorch_model-00002-of-00004.safetensors",
|
| 408 |
+
"layers.25.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 409 |
+
"layers.25.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 410 |
+
"layers.25.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 411 |
+
"layers.25.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 412 |
+
"layers.25.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 413 |
+
"layers.25.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 414 |
+
"layers.25.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 415 |
+
"layers.25.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 416 |
+
"layers.25.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 417 |
+
"layers.25.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 418 |
+
"layers.25.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 419 |
+
"layers.25.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 420 |
+
"layers.25.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 421 |
+
"layers.25.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 422 |
+
"layers.25.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 423 |
+
"layers.25.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 424 |
+
"layers.25.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 425 |
+
"layers.25.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 426 |
+
"layers.25.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 427 |
+
"layers.25.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 428 |
+
"layers.25.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 429 |
+
"layers.25.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 430 |
+
"layers.26.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 431 |
+
"layers.26.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 432 |
+
"layers.26.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 433 |
+
"layers.26.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 434 |
+
"layers.26.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 435 |
+
"layers.26.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 436 |
+
"layers.26.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 437 |
+
"layers.26.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 438 |
+
"layers.26.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 439 |
+
"layers.26.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 440 |
+
"layers.26.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 441 |
+
"layers.26.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 442 |
+
"layers.26.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 443 |
+
"layers.26.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 444 |
+
"layers.26.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 445 |
+
"layers.26.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 446 |
+
"layers.26.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 447 |
+
"layers.26.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 448 |
+
"layers.26.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 449 |
+
"layers.26.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 450 |
+
"layers.26.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 451 |
+
"layers.26.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 452 |
+
"layers.27.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 453 |
+
"layers.27.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 454 |
+
"layers.27.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 455 |
+
"layers.27.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 456 |
+
"layers.27.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 457 |
+
"layers.27.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 458 |
+
"layers.27.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 459 |
+
"layers.27.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 460 |
+
"layers.27.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 461 |
+
"layers.27.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 462 |
+
"layers.27.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 463 |
+
"layers.27.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 464 |
+
"layers.27.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 465 |
+
"layers.27.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 466 |
+
"layers.27.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 467 |
+
"layers.27.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 468 |
+
"layers.27.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 469 |
+
"layers.27.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 470 |
+
"layers.27.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 471 |
+
"layers.27.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 472 |
+
"layers.27.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 473 |
+
"layers.27.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 474 |
+
"layers.28.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 475 |
+
"layers.28.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 476 |
+
"layers.28.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 477 |
+
"layers.28.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 478 |
+
"layers.28.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 479 |
+
"layers.28.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 480 |
+
"layers.28.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 481 |
+
"layers.28.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 482 |
+
"layers.28.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 483 |
+
"layers.28.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 484 |
+
"layers.28.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 485 |
+
"layers.28.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 486 |
+
"layers.28.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 487 |
+
"layers.28.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 488 |
+
"layers.28.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 489 |
+
"layers.28.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 490 |
+
"layers.28.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 491 |
+
"layers.28.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 492 |
+
"layers.28.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 493 |
+
"layers.28.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 494 |
+
"layers.28.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 495 |
+
"layers.28.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 496 |
+
"layers.29.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 497 |
+
"layers.29.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 498 |
+
"layers.29.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 499 |
+
"layers.29.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 500 |
+
"layers.29.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 501 |
+
"layers.29.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 502 |
+
"layers.29.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 503 |
+
"layers.29.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 504 |
+
"layers.29.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 505 |
+
"layers.29.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 506 |
+
"layers.29.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 507 |
+
"layers.29.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 508 |
+
"layers.29.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 509 |
+
"layers.29.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 510 |
+
"layers.29.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 511 |
+
"layers.29.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 512 |
+
"layers.29.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 513 |
+
"layers.29.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 514 |
+
"layers.29.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 515 |
+
"layers.29.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 516 |
+
"layers.29.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 517 |
+
"layers.29.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 518 |
+
"layers.3.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 519 |
+
"layers.3.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 520 |
+
"layers.3.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 521 |
+
"layers.3.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 522 |
+
"layers.3.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 523 |
+
"layers.3.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 524 |
+
"layers.3.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 525 |
+
"layers.3.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 526 |
+
"layers.3.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 527 |
+
"layers.3.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 528 |
+
"layers.3.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 529 |
+
"layers.3.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 530 |
+
"layers.3.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 531 |
+
"layers.3.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 532 |
+
"layers.3.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 533 |
+
"layers.3.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 534 |
+
"layers.3.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 535 |
+
"layers.3.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 536 |
+
"layers.3.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 537 |
+
"layers.3.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 538 |
+
"layers.3.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 539 |
+
"layers.3.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 540 |
+
"layers.30.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 541 |
+
"layers.30.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 542 |
+
"layers.30.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 543 |
+
"layers.30.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 544 |
+
"layers.30.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 545 |
+
"layers.30.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 546 |
+
"layers.30.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 547 |
+
"layers.30.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 548 |
+
"layers.30.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 549 |
+
"layers.30.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 550 |
+
"layers.30.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 551 |
+
"layers.30.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 552 |
+
"layers.30.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 553 |
+
"layers.30.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 554 |
+
"layers.30.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 555 |
+
"layers.30.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 556 |
+
"layers.30.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 557 |
+
"layers.30.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 558 |
+
"layers.30.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 559 |
+
"layers.30.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 560 |
+
"layers.30.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 561 |
+
"layers.30.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 562 |
+
"layers.31.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 563 |
+
"layers.31.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 564 |
+
"layers.31.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 565 |
+
"layers.31.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 566 |
+
"layers.31.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 567 |
+
"layers.31.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 568 |
+
"layers.31.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 569 |
+
"layers.31.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 570 |
+
"layers.31.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 571 |
+
"layers.31.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 572 |
+
"layers.31.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 573 |
+
"layers.31.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 574 |
+
"layers.31.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 575 |
+
"layers.31.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 576 |
+
"layers.31.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 577 |
+
"layers.31.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 578 |
+
"layers.31.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 579 |
+
"layers.31.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 580 |
+
"layers.31.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 581 |
+
"layers.31.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 582 |
+
"layers.31.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 583 |
+
"layers.31.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 584 |
+
"layers.32.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 585 |
+
"layers.32.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 586 |
+
"layers.32.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 587 |
+
"layers.32.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 588 |
+
"layers.32.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 589 |
+
"layers.32.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 590 |
+
"layers.32.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 591 |
+
"layers.32.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 592 |
+
"layers.32.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 593 |
+
"layers.32.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 594 |
+
"layers.32.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 595 |
+
"layers.32.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 596 |
+
"layers.32.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 597 |
+
"layers.32.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 598 |
+
"layers.32.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 599 |
+
"layers.32.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 600 |
+
"layers.32.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 601 |
+
"layers.32.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 602 |
+
"layers.32.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 603 |
+
"layers.32.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 604 |
+
"layers.32.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 605 |
+
"layers.32.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 606 |
+
"layers.33.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 607 |
+
"layers.33.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 608 |
+
"layers.33.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 609 |
+
"layers.33.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 610 |
+
"layers.33.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 611 |
+
"layers.33.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 612 |
+
"layers.33.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 613 |
+
"layers.33.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 614 |
+
"layers.33.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 615 |
+
"layers.33.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 616 |
+
"layers.33.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 617 |
+
"layers.33.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 618 |
+
"layers.33.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 619 |
+
"layers.33.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 620 |
+
"layers.33.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 621 |
+
"layers.33.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 622 |
+
"layers.33.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 623 |
+
"layers.33.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 624 |
+
"layers.33.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 625 |
+
"layers.33.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 626 |
+
"layers.33.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 627 |
+
"layers.33.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 628 |
+
"layers.34.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 629 |
+
"layers.34.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 630 |
+
"layers.34.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 631 |
+
"layers.34.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 632 |
+
"layers.34.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 633 |
+
"layers.34.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 634 |
+
"layers.34.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 635 |
+
"layers.34.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 636 |
+
"layers.34.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 637 |
+
"layers.34.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 638 |
+
"layers.34.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 639 |
+
"layers.34.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 640 |
+
"layers.34.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 641 |
+
"layers.34.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 642 |
+
"layers.34.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 643 |
+
"layers.34.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 644 |
+
"layers.34.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 645 |
+
"layers.34.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 646 |
+
"layers.34.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 647 |
+
"layers.34.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 648 |
+
"layers.34.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 649 |
+
"layers.34.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 650 |
+
"layers.35.input_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 651 |
+
"layers.35.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 652 |
+
"layers.35.mlp.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 653 |
+
"layers.35.mlp.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 654 |
+
"layers.35.mlp.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 655 |
+
"layers.35.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 656 |
+
"layers.35.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 657 |
+
"layers.35.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 658 |
+
"layers.35.post_attention_layernorm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 659 |
+
"layers.35.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 660 |
+
"layers.35.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 661 |
+
"layers.35.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 662 |
+
"layers.35.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 663 |
+
"layers.35.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 664 |
+
"layers.35.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 665 |
+
"layers.35.self_attn.norm_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 666 |
+
"layers.35.self_attn.norm_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 667 |
+
"layers.35.self_attn.to_add_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 668 |
+
"layers.35.self_attn.to_k.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 669 |
+
"layers.35.self_attn.to_out.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 670 |
+
"layers.35.self_attn.to_q.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 671 |
+
"layers.35.self_attn.to_v.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 672 |
+
"layers.4.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 673 |
+
"layers.4.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 674 |
+
"layers.4.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 675 |
+
"layers.4.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 676 |
+
"layers.4.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 677 |
+
"layers.4.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 678 |
+
"layers.4.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 679 |
+
"layers.4.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 680 |
+
"layers.4.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 681 |
+
"layers.4.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 682 |
+
"layers.4.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 683 |
+
"layers.4.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 684 |
+
"layers.4.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 685 |
+
"layers.4.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 686 |
+
"layers.4.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 687 |
+
"layers.4.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 688 |
+
"layers.4.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 689 |
+
"layers.4.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 690 |
+
"layers.4.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 691 |
+
"layers.4.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 692 |
+
"layers.4.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 693 |
+
"layers.4.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 694 |
+
"layers.5.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 695 |
+
"layers.5.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 696 |
+
"layers.5.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 697 |
+
"layers.5.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 698 |
+
"layers.5.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 699 |
+
"layers.5.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 700 |
+
"layers.5.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 701 |
+
"layers.5.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 702 |
+
"layers.5.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 703 |
+
"layers.5.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 704 |
+
"layers.5.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 705 |
+
"layers.5.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 706 |
+
"layers.5.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 707 |
+
"layers.5.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 708 |
+
"layers.5.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 709 |
+
"layers.5.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 710 |
+
"layers.5.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 711 |
+
"layers.5.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 712 |
+
"layers.5.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 713 |
+
"layers.5.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 714 |
+
"layers.5.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 715 |
+
"layers.5.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 716 |
+
"layers.6.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 717 |
+
"layers.6.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 718 |
+
"layers.6.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 719 |
+
"layers.6.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 720 |
+
"layers.6.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 721 |
+
"layers.6.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 722 |
+
"layers.6.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 723 |
+
"layers.6.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 724 |
+
"layers.6.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 725 |
+
"layers.6.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 726 |
+
"layers.6.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 727 |
+
"layers.6.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 728 |
+
"layers.6.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 729 |
+
"layers.6.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 730 |
+
"layers.6.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 731 |
+
"layers.6.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 732 |
+
"layers.6.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 733 |
+
"layers.6.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 734 |
+
"layers.6.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 735 |
+
"layers.6.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 736 |
+
"layers.6.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 737 |
+
"layers.6.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 738 |
+
"layers.7.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 739 |
+
"layers.7.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 740 |
+
"layers.7.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 741 |
+
"layers.7.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 742 |
+
"layers.7.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 743 |
+
"layers.7.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 744 |
+
"layers.7.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 745 |
+
"layers.7.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 746 |
+
"layers.7.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 747 |
+
"layers.7.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 748 |
+
"layers.7.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 749 |
+
"layers.7.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 750 |
+
"layers.7.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 751 |
+
"layers.7.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 752 |
+
"layers.7.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 753 |
+
"layers.7.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 754 |
+
"layers.7.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 755 |
+
"layers.7.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 756 |
+
"layers.7.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 757 |
+
"layers.7.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 758 |
+
"layers.7.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 759 |
+
"layers.7.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 760 |
+
"layers.8.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 761 |
+
"layers.8.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 762 |
+
"layers.8.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 763 |
+
"layers.8.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 764 |
+
"layers.8.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 765 |
+
"layers.8.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 766 |
+
"layers.8.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 767 |
+
"layers.8.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 768 |
+
"layers.8.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 769 |
+
"layers.8.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 770 |
+
"layers.8.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 771 |
+
"layers.8.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 772 |
+
"layers.8.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 773 |
+
"layers.8.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 774 |
+
"layers.8.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 775 |
+
"layers.8.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 776 |
+
"layers.8.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 777 |
+
"layers.8.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 778 |
+
"layers.8.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 779 |
+
"layers.8.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 780 |
+
"layers.8.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 781 |
+
"layers.8.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 782 |
+
"layers.9.input_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 783 |
+
"layers.9.input_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 784 |
+
"layers.9.mlp.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 785 |
+
"layers.9.mlp.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 786 |
+
"layers.9.mlp.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 787 |
+
"layers.9.mlp_moe_gen.down_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 788 |
+
"layers.9.mlp_moe_gen.gate_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 789 |
+
"layers.9.mlp_moe_gen.up_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 790 |
+
"layers.9.post_attention_layernorm.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 791 |
+
"layers.9.post_attention_layernorm_moe_gen.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 792 |
+
"layers.9.self_attn.add_k_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 793 |
+
"layers.9.self_attn.add_q_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 794 |
+
"layers.9.self_attn.add_v_proj.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 795 |
+
"layers.9.self_attn.norm_added_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 796 |
+
"layers.9.self_attn.norm_added_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 797 |
+
"layers.9.self_attn.norm_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 798 |
+
"layers.9.self_attn.norm_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 799 |
+
"layers.9.self_attn.to_add_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 800 |
+
"layers.9.self_attn.to_k.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 801 |
+
"layers.9.self_attn.to_out.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 802 |
+
"layers.9.self_attn.to_q.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 803 |
+
"layers.9.self_attn.to_v.weight": "diffusion_pytorch_model-00001-of-00004.safetensors",
|
| 804 |
+
"lm_head.weight": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 805 |
+
"norm.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 806 |
+
"norm_moe_gen.weight": "diffusion_pytorch_model-00003-of-00004.safetensors",
|
| 807 |
+
"proj_in.bias": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 808 |
+
"proj_in.weight": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 809 |
+
"proj_out.bias": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 810 |
+
"proj_out.weight": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 811 |
+
"time_embedder.linear_1.bias": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 812 |
+
"time_embedder.linear_1.weight": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 813 |
+
"time_embedder.linear_2.bias": "diffusion_pytorch_model-00004-of-00004.safetensors",
|
| 814 |
+
"time_embedder.linear_2.weight": "diffusion_pytorch_model-00004-of-00004.safetensors"
|
| 815 |
+
}
|
| 816 |
+
}
|
vae/config.json
ADDED
|
@@ -0,0 +1,129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "AutoencoderKLWan",
|
| 3 |
+
"_diffusers_version": "0.39.0.dev0",
|
| 4 |
+
"_name_or_path": "/workspace/cosmos3-nano/vae",
|
| 5 |
+
"attn_scales": [],
|
| 6 |
+
"base_dim": 160,
|
| 7 |
+
"clip_output": false,
|
| 8 |
+
"decoder_base_dim": 256,
|
| 9 |
+
"dim_mult": [
|
| 10 |
+
1,
|
| 11 |
+
2,
|
| 12 |
+
4,
|
| 13 |
+
4
|
| 14 |
+
],
|
| 15 |
+
"dropout": 0.0,
|
| 16 |
+
"in_channels": 12,
|
| 17 |
+
"is_residual": true,
|
| 18 |
+
"latents_mean": [
|
| 19 |
+
-0.2289,
|
| 20 |
+
-0.0052,
|
| 21 |
+
-0.1323,
|
| 22 |
+
-0.2339,
|
| 23 |
+
-0.2799,
|
| 24 |
+
0.0174,
|
| 25 |
+
0.1838,
|
| 26 |
+
0.1557,
|
| 27 |
+
-0.1382,
|
| 28 |
+
0.0542,
|
| 29 |
+
0.2813,
|
| 30 |
+
0.0891,
|
| 31 |
+
0.157,
|
| 32 |
+
-0.0098,
|
| 33 |
+
0.0375,
|
| 34 |
+
-0.1825,
|
| 35 |
+
-0.2246,
|
| 36 |
+
-0.1207,
|
| 37 |
+
-0.0698,
|
| 38 |
+
0.5109,
|
| 39 |
+
0.2665,
|
| 40 |
+
-0.2108,
|
| 41 |
+
-0.2158,
|
| 42 |
+
0.2502,
|
| 43 |
+
-0.2055,
|
| 44 |
+
-0.0322,
|
| 45 |
+
0.1109,
|
| 46 |
+
0.1567,
|
| 47 |
+
-0.0729,
|
| 48 |
+
0.0899,
|
| 49 |
+
-0.2799,
|
| 50 |
+
-0.123,
|
| 51 |
+
-0.0313,
|
| 52 |
+
-0.1649,
|
| 53 |
+
0.0117,
|
| 54 |
+
0.0723,
|
| 55 |
+
-0.2839,
|
| 56 |
+
-0.2083,
|
| 57 |
+
-0.052,
|
| 58 |
+
0.3748,
|
| 59 |
+
0.0152,
|
| 60 |
+
0.1957,
|
| 61 |
+
0.1433,
|
| 62 |
+
-0.2944,
|
| 63 |
+
0.3573,
|
| 64 |
+
-0.0548,
|
| 65 |
+
-0.1681,
|
| 66 |
+
-0.0667
|
| 67 |
+
],
|
| 68 |
+
"latents_std": [
|
| 69 |
+
0.4765,
|
| 70 |
+
1.0364,
|
| 71 |
+
0.4514,
|
| 72 |
+
1.1677,
|
| 73 |
+
0.5313,
|
| 74 |
+
0.499,
|
| 75 |
+
0.4818,
|
| 76 |
+
0.5013,
|
| 77 |
+
0.8158,
|
| 78 |
+
1.0344,
|
| 79 |
+
0.5894,
|
| 80 |
+
1.0901,
|
| 81 |
+
0.6885,
|
| 82 |
+
0.6165,
|
| 83 |
+
0.8454,
|
| 84 |
+
0.4978,
|
| 85 |
+
0.5759,
|
| 86 |
+
0.3523,
|
| 87 |
+
0.7135,
|
| 88 |
+
0.6804,
|
| 89 |
+
0.5833,
|
| 90 |
+
1.4146,
|
| 91 |
+
0.8986,
|
| 92 |
+
0.5659,
|
| 93 |
+
0.7069,
|
| 94 |
+
0.5338,
|
| 95 |
+
0.4889,
|
| 96 |
+
0.4917,
|
| 97 |
+
0.4069,
|
| 98 |
+
0.4999,
|
| 99 |
+
0.6866,
|
| 100 |
+
0.4093,
|
| 101 |
+
0.5709,
|
| 102 |
+
0.6065,
|
| 103 |
+
0.6415,
|
| 104 |
+
0.4944,
|
| 105 |
+
0.5726,
|
| 106 |
+
1.2042,
|
| 107 |
+
0.5458,
|
| 108 |
+
1.6887,
|
| 109 |
+
0.3971,
|
| 110 |
+
1.06,
|
| 111 |
+
0.3943,
|
| 112 |
+
0.5537,
|
| 113 |
+
0.5444,
|
| 114 |
+
0.4089,
|
| 115 |
+
0.7468,
|
| 116 |
+
0.7744
|
| 117 |
+
],
|
| 118 |
+
"num_res_blocks": 2,
|
| 119 |
+
"out_channels": 12,
|
| 120 |
+
"patch_size": 2,
|
| 121 |
+
"scale_factor_spatial": 16,
|
| 122 |
+
"scale_factor_temporal": 4,
|
| 123 |
+
"temperal_downsample": [
|
| 124 |
+
false,
|
| 125 |
+
true,
|
| 126 |
+
true
|
| 127 |
+
],
|
| 128 |
+
"z_dim": 48
|
| 129 |
+
}
|
vae/diffusion_pytorch_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:230496cb59ff85bc9c040487737c4062480cb61c71e697b197b4c30142f2a0da
|
| 3 |
+
size 1409400600
|